1. Sukhvinder S Bamber and Ajay K Sharma
International Journal of Computer Networks (IJCN), Volume (2): Issue (4) 181
Queue Size Trade Off with Modulation in 802.15.4 for Wireless
Sensor Networks
Sukhvinder S Bamber bambery2k@yahoo.com
Department of Computer Science & Engineering
National Institute of Technology,
Jalandhar, 144011, India
Ajay K Sharma sharmaajayk@nitj.ac.in
Department of Computer Science & Engineering
National Institute of Technology,
Jalandhar, 144011, India
Abstract
In this paper we analyze the performance of 802.15.4 Wireless Sensor Network
(WSN) and derive the queue size trade off for different modulation schemes like:
Minimum Shift Keying (MSK), Quadrature Amplitude Modulation of 64 bits
(QAM_64) and Binary Phase Shift Keying (BPSK) at the radio transmitter of
different types of devices in IEEE 802.15.4 for WSN. It is concluded that if queue
size at the PAN coordinator of 802.15.4 wireless sensor network is to be taken
into consideration then QAM_64 is recommended. Also it has been concluded
that if the queue size at the GTS or Non GTS end device is to be considered then
BPSK should be preferred. Our results can be used for planning and deploying
IEEE 802.15.4 based wireless sensor networks with specific performance
demands. Overall it has been revealed that there is trade off for using various
modulation schemes in WSN devices.
Keywords: WSN, Queue Size, BPSK, MSK, QAM_64.
1. INTRODUCTION
The IEEE 802.15.4 protocol is an industrial standard for Low-Rate Wireless Personal Area
Network (LR-WPAN) architectures. As the primary application domain wireless sensor network
applications in industrial environments can be identified. LR-WPAN is intended to become an
enabling technology for WSNs. In contrast to Wireless Local Area Networks (WLAN), which is
standardized by IEEE 802.11 family, LR-WPAN stresses short-range operation, low-data-rate,
energy-efficiency and low-cost. An example is Zigbee, which is an open specification built on the
LR-WPAN standard and focuses on the establishment and maintenance of LR-WPANs for
wireless sensor networks.
The choice of the digital modulation scheme significantly affects the characteristics, performance
and resulting physical realization of wireless sensor communication system derived from
802.15.4. There is no universal ‘best’ choice of the modulation scheme, but depending on the
physical characteristics of the channel, parametric optimizations and required level of
performance some will prove better fit than the others. The 802.15.4 is an IEEE standard,

2. Sukhvinder S Bamber and Ajay K Sharma
International Journal of Computer Networks (IJCN), Volume (2): Issue (4) 182
targeting a set of applications that require simple wireless connectivity, high throughput, very low
power consumption and lower module cost. Its objective is to provide low complexity, cost and
power for wireless sensor connectivity among inexpensive, fixed, portable and moving devices.
A lot of work on 802.15.4 has been reported by the various researchers [1-22]. The performance
issues like: Delay; Throughput evaluation of GTS mechanism have been reported in [1].
Researchers have also studied adaptive algorithm for mapping channel quality information to
modulation and coding schemes [3]. Researchers have also tried to study performance tradeoff
with adaptive frame length and modulation in wireless network [4]. Some researchers have
studied suboptimum receivers for DS-CDMA with BPSK modulation [5]. Researchers have also
investigated voice and data transmission technique using adaptive modulation [6]. Many
researchers have studied how to use queues to improve the performance of TCP [10]. Some
have studied queues o dynamically allocate the channels for real-time and non-real-time traffic in
cellular networks [12]. Few have studied queues for energy and QoS tradeoff for contention-
based wireless sensor networks [15]. Some have worked on how to stabilize queues in large-
scale networks [16]. Few researchers have studied the queues for controlling the power in
wireless communication networks [20]. But none of the researchers have reported the
performance comparison using different modulation schemes for 802.15.4 based on queue size.
This paper proposes the comparison of different modulation schemes (QAM_64, MSK, BPSK)
based on queue size to determine the suitability of 802.15.4 network.
Section [1] gives the brief introduction. Section [2] constitutes the system description which
contains node model, process model, and parametric tables of the model. Section [3] shows the
results and discussions derived from the experiments carried out on 802.15.4 for different
modulation schemes. Finally Section [4] concludes the paper.
2. SYSTEM DESCRIPTION
The simulation model implements physical and medium access layers defined in IEEE 802.15.4
standard. The OPNET
®
Modeler 14.5 is used for developing 802.15.4 wireless sensor network.
(a) (b)

3. Sukhvinder S Bamber and Ajay K Sharma
International Journal of Computer Networks (IJCN), Volume (2): Issue (4) 183
(c)
Figure 1: Network Scenarios (a) BPSK (b) MSK (c) Quadrature (QAM_64)
Figure 1 shows three different Scenarios: BPSK, MSK and QAM_64. BPSK Scenario as shown in
Figure 1(a) contains one PAN Coordinator, one analyzer and thirty two end devices out of which
sixteen are Guaranteed Time Slots (GTS) enabled and rest are non GTS devices. PAN
Coordinator is a fully functional device which manages whole functioning of the network. Analyzer
is a routing device which routes the data between PAN coordinator and the End Devices. End
Devices are the fixed stations that communicate with the PAN Coordinator in Peer to Peer mode,
support GTS and non GTS traffic respectively. Similar Scenarios have been created for MSK and
QAM_64 as shown in figure 1 (b & c).
Figure 2 shows the node models for three types of WPAN devices used for modeling 802.15.4
scenarios. PAN Coordinator, GTS and Non GTS end device have the same node model as
shown in Figure 2 (a) while the node model for analyzer is depicted in Figure 2 (b).
(a) (b)
Figure 2: Node Model (a) PAN Coordinator, GTS and Non GTS end device (b) Analyzer

4. Sukhvinder S Bamber and Ajay K Sharma
International Journal of Computer Networks (IJCN), Volume (2): Issue (4) 184
As it has been observed from the Figure 2 (a), a node model for PAN Coordinator, GTS end
device and Non GTS end device has three layers: physical, MAC and application layers. Physical
layer consists of a transmitter and a receiver compliant to the IEEE 802.15.4 specification,
operating at 2.4 GHz frequency band and data rate equal to 250 kbps. MAC layer implements
slotted CSMA/CA and GTS mechanisms. The GTS data traffic coming from the application layer
is stored in a buffer with a specified capacity and dispatched to the network when the
corresponding GTS is active. The non time-critical data frames are stored in an unbounded buffer
and based on slotted CSMA/CA algorithm are transmitted to the network during the active
Contention Access Period (CAP). This layer is also responsible for the generation of beacon
frames and synchronizing the network when a given node acts as a PAN Coordinator. Finally is
the topmost application layer which is responsible for generation and reception of traffic consists
of two data traffic generators (i.e. Traffic Source and GTS Traffic Source) and one traffic sink. The
traffic source generates acknowledged and unacknowledged data frames transmitted during
CAP. GTS traffic source can produce acknowledged and unacknowledged time-critical data
frames using GTS mechanism. The traffic sink module receives frames forwarded from lower
layers. Figure 2 (b) shows the node model for the analyzer which consists of sink and a radio
receiver.
Corresponding process models for PAN Coordinator, GTS end device, Non GTS end device and
analyzer that deals with each and every operation on the data are depicted in Figure 3:
(a)
(b)
Figure 3: Process model (a) PAN Coordinator, GTS and Non GTS end device (b) Analyzer
Figure 3 (a) shows the process model for the PAN Coordinator, GTS and Non GTS end device. It
consists of the various states: Init whose function is to initialize MAC and GTS scheduling;

5. Sukhvinder S Bamber and Ajay K Sharma
International Journal of Computer Networks (IJCN), Volume (2): Issue (4) 185
Wait_beacon which is responsible for synchronizing the traffic of the node with rest of the WPAN
in order to minimize the collisions; Idle which is responsible for introducing delays in order to
make the maximum use of the resources; gts_slot which is responsible for generation, reception
and management of GTS traffic; Backoff_timer used for sensing the medium and transfer of data,
CCA - for interrupt processing. Similarly figure 3 (b) shows the process model for analyzer which
consists of init and idle states. Basically the process model explains how the data is sent from the
generating node to the PAN Coordinator, taking into consideration the availability of PAN
Coordinator as it has to communicate with the other similar nodes.
Here three different Scenarios have been created with three different modulation formats like:
BPSK, MSK and QAM_64. Following parameters have been set for these scenarios as shown in
the table 1 like: in GTS settings the value of GTS permit is common for all three types of devices
i.e. enabled.
Parameter Scenario PAN
Coordinator
GTS Enabled
End Device
Non GTS
End Device
Modulation BPSK, MSK, QAM_64
Acknowledged Traffic Source
Destination MAC Address Broadcast PAN Coordinator
MSDU Interarrival Time
(sec)
Exponential(1.0) Constant (1.0) Exponential(1.0)
MSDU Size (bits) Exponential(912) Constant (0.0) Exponential(912)
Start Time (sec) 0.0 Infinity 1.0
Stop Time (sec) Infinity
Unacknowledged Traffic Source
MSDU Interarrival Time
(sec)
Exponential(1.0) Constant (1.0) Exponential(1.0)
MSDU Size (bits) Exponential(912) Constant (0.0) Exponential(912)
Start Time (sec) 0.1 Infinity 1.1
Stop Time (sec) Infinity
CSMA/CA Parameters
Maximum Back-off Number 4
Minimum Back-off
Exponent
3
IEEE 802.15.4
Device Mode PAN coordinator End Device
MAC Address Auto Assigned
WPAN Settings
Beacon Order 14 7
Superframe Order 6
PAN ID 0
Logging
Enable Logging Enabled
GTS Settings
GTS Permit Enabled
Start Time 0.0 0.1 Infinity
Stop Time Infinity
Length (slots) 1 0
Direction Receive Transmit
Buffer Capacity (bits) 10,000 1000
GTS Traffic Parameters
MSDU Interarrival Time
(sec)
Exponential(1.0) Constant (1.0)
MSDU Size (bits) Exponential(912) Constant (0.0)
Acknowledgement Enabled Disabled
Table 1: Parametric values for PAN Coordinator, GTS and Non GTS End Device in BPSK, MSK and
QAM_64 Scenarios

6. Sukhvinder S Bamber and Ajay K Sharma
International Journal of Computer Networks (IJCN), Volume (2): Issue (4) 186
3. RESULTS AND DISCUSSIONS
Simulation has been carried out for the three different scenarios of IEEE 802.15.4 using QAM_64,
MSK and BPSK. In this section results for the queue size at the radio transmitter have been
presented and discussed for different types of devices in wireless sensor networks like: Fully
Functional Device (FFD) – those devices that control the network and manage the routing tables
and communicate with each of the device in peer to peer mode, Reduced Functional Devices
(RFD) – those devices which can only communicate to the FFD but not to each other.
Radio Transmitter Queue Size
3.1.1 FFD – PAN Coordinator
Figure 4 below indicates the queue size at the radio transmitter of a PAN Coordinator. It is
observed that it is 0.2926, 0.2572 and 0.2261 packets for MSK, BPSK and QAM_64 respectively.
It has been experimentally proved that queue size is maximum in case of MSK because it
purposefully generates the delays to reduce the phase shifts to produce amplifier-friendly signals
which results in the long queues at the radio transmitter as compared to the other modulation
schemes (e.g. BPSK, QAM_64 etc.) and also MSK has self synchronizing capability [17]. While it
has been observed that queue size is minimum in case of QAM_64 as it increases the efficiency
of transmission by utilizing both amplitude and phase variations [17, 23].
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0 132 264 396 528 660 792 924 1056 1188
Time (sec)
RadioTransmitterQueueSize
(packets)
BPSK MSK QAM_64
Figure 4: Radio Transmitter Queue Size at PAN Coordinator
3.1.2 RFD – GTS End Device
Figure 5 indicates the queue size at the radio transmitter of a GTS end device. It is 0.0179,
0.0020 and 0.0013 packets MSK, QAM_64 and BPSK respectively. It has been observed that
queue size is maximum in case of MSK [17]. While it is minimum in case of BPSK as it can
modulate only 01 bit/sec and there is strong synchronization between the transmitter and the
receiver [23].

7. Sukhvinder S Bamber and Ajay K Sharma
International Journal of Computer Networks (IJCN), Volume (2): Issue (4) 187
0
0.002
0.004
0.006
0.008
0.01
0.012
0.014
0.016
0.018
0.02
0 132 264 396 528 660 792 924 1056 1188
Time (sec)
RadioTransmitterQueueSize
(packets)
BPSK MSK QAM_64
Figure 5: Radio Transmitter Queue Size at GTS End Device
3.1.3 RFD – Non GTS End Device
Figure 6 reveals the queue size at the radio transmitter of a Non GTS end device. It is 0.1621,
0.1340 and 0.1172 packets for MSK, QAM_64 and BPSK respectively. It has been observed that
it is maximum in case of MSK [17], while it is minimum in case of BPSK [23].
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
0 132 264 396 528 660 792 924 1056 1188
Time (sec)
RadioTransmitterQueueSize
(packets)
BPSK MSK QAM_64
Figure 6: Radio Transmitter Queue Size at Non GTS End Device
From the results obtained in figures: 4 for FFD and 5 & 6 for RFD (GTS & Non GTS), it has been
concluded that if queue size at the PAN coordinator of 802.15.4 wireless sensor network is to be
taken into consideration then QAM_64 should be preferred and if queue size at the GTS or Non
GTS end device is to be considered then BPSK should be preferred.
4. CONSLUSION
This paper presents the queue size at the radio transmitter of 802.15.4 wireless sensor network
using OPNET
®
Modeler 14.5. Here three different modulation scenarios for BPSK, MSK and
QAM_64 have been considered. Results reveals that queue size at the radio transmitter of PAN
Coordinator, GTS and Non GTS End Device is [0.2926, 0.2261, 0.2572], [0.0179, 0.0020, 0.0013]
and [0.1621, 0.1340, 0.1172] packets for MSK, QAM_64 and BPSK respectively. It is concluded
that QAM_64 at the fully functional device and BPSK at the GTS and Non GTS RFDs should be
implemented if queue size at the radio transmitter of 802.15.4 WSN is to be minimized. Also it is

8. Sukhvinder S Bamber and Ajay K Sharma
International Journal of Computer Networks (IJCN), Volume (2): Issue (4) 188
concluded that MSK at all type of devices in 802.15.4 for WSN is unsuitable as it results in the
larger queues as compared to the other modulation formats at all type of devices, as larger the
queues, larger will be the delays.
5. REFERENCES
[1] Jurcik, P., Koubaa, A., Alves, M., Tovar, E., Hanzalek, Z. “A Simulation Model for the
IEEE 802.15.4 protocol: Delay/Throughput Evaluation of the GTS Mechanism”. Modeling,
Analysis, and Simulation of Computer and Telecommunication Systems, 2007:
MASCOTS '07. 15th International Symposium, 24-26 Oct. 2007.
[2] IEEE 802.15.4 OPNET Simulation Model: http://www.open-zb.net.
[3] Patrick Hosein. “Adaptive Algorithm for Mapping Channel Quality Information to
Modulation and Coding Schemes”. IEEE: 2009.
[4] Yafei Hou, Masanori Hamamura, Shiyong Zhang. “Performance Tradeoff with Adaptive
Frame Length and Modulation in wireless Network”. Proceedings of the IEEE 2005, the
Fifth International Conference on Computer and Information Technology (CIT’ 05).
[5] R. Schober, W. H. Gerstacker, L. Lampe. “On suboptimum receivers for DS-CDMA with
BPSK modulation”. Signal Processing 85 (2005): 1149 – 1163, Elsevier.
[6] Rajarshi Mahapatra, Anindya Sunder Char, Debasish Datta. “Dynamic Capacity
Allocation for Voice and Data Using Adaptive Modulation in Wireless Networks”. IEEE:
2006.
[7] Jan Magne Tjensvold. “Comparision of the IEEE 802.11, 802.15.1, 802.15.4 and
802.15.6 wireless standards”. IEEE: September 18, 2007.
[8] Jason Lowe, “Advanced Upstream Modulation”. Clearcable Technical Summit: July 27,
2007.
[9] Feng Chen, Nan Wang, Reinhard German, Falko Dressler. “Simulation study of IEEE
802.15.4 LR-WPAN for industrial applications”. Wireless Communications and Mobile
Computing: 2009.
[10] S.M. Mahdi Alavi, Martin J. Hayes. “Robust Active Queue management design: A loop-
shaping approach”. Elsevier: Computer Communications, 2009.
[11] Shan Chen, Brahim Bensaaou. “Can high-speed networks survive with DropTail queue
management”. Elsevier: Computer Networks, 2006.
[12] P. Venkata Krishna, Sudip Misra, Mohammad S. Obaidat, V. Saritha. “An efficient
approach for distributed dynamic channel allocation with queues for real-time and non-
real-time traffic in cellular networks”. Elsevier: The Journal of Systems and Software,
2009.
[13] Subhash Nanjunde Gowda. “Minimum shift keying”. Spread Spectrum Systems: 24 May
2004.
[14] Qiuyan Xia, Xing Jin, Mounir Hamdi. “AQM with Dual Virtual PI Queues for TCP
Uplink/Downlink Fairness in Infrastructure WLANs”. IEEE: 2007.
[15] Jun Luo, Lingge Jiang, Chen He, “Finite Queuing Model Analysis for Energy and QoS
Tradeoff in Contention-Based Wireless Sensor Networks”, IEEE, 2007.
[16] Yi Fan, Zhong-Ping Jiang, Hao Zhang. “Stablizing Queues in Large-scale Networks”.
IEEE: 2005.
[17] Charan Langton. “Intuitive Guide to principles of Communications”.
www.complextoreal.com – Dec 2005.
[18] Prasan Kumar Sahoo, Jang-Ping Sheu. “Modelling IEEE 802.15.4 based Wireless
Sensor Network with Packet Retry Limits”. ACM: PE-WASUN’08.
[19] Jelena Misic, Vojislav B. Misic. “Queuing Analysis of Sleep Management in an 802.15.4
Beacon Enabled PAN”. IEEE.
[20] L. Chisci, R. Fantacci, L. Mucchi, T. Pecorella. “A Queue-Based Approach to Power
Control in Wireless Communication Networks”. IEEE: 2008.
[21] Shahram Teymori, Weihua Zhuang. “Finite Buffer Queue Analysis and Scheduling for
Heavy-tailed Traffic in Packet-Switching wireless Networks”. IEEE: 2005.

9. Sukhvinder S Bamber and Ajay K Sharma
International Journal of Computer Networks (IJCN), Volume (2): Issue (4) 189
[22] Omesh Tickoo, Biplab Sikdar. “Queuing Analysis and Delay Mitigation in IEEE 802.11
Random access MAC based Wireless Networks”. IEEE: 2004.
[23] http://www.en.wikipedia.org.